Methods, devices, equipment and storage media for detecting defects in ballastless track in regional areas

Abstract

This invention relates to the field of track defect detection technology, and discloses a method, apparatus, equipment, and storage medium for regional ballastless track defect detection. Based on a historical defect database and communication quality map of ballastless track, it generates the data upload requirements and capabilities for each inspection sub-segment. Then, using the speed of the inspection vehicle as the core, combined with hardware constraints and gear shifting optimization objectives, it solves for the optimal detection strategy, thereby guiding the execution of ballastless track inspection driving, data upload, and cloud-based defect detection analysis. Therefore, by optimizing the inspection speed of the ballastless track inspection vehicle in different inspection sub-segments, this invention ensures that inspections are completed within the maintenance window while completely resolving the problem of data accumulation in areas with poor network connectivity, such as tunnels and mountainous regions, leading to inspection stoppages or the overwriting of old data. This guarantees the integrity of the inspection task and the reliability of the data, helps the cloud quickly analyze and output defect detection results, and provides timely information for track maintenance.

Description

Technical Field

[0001] This invention relates to the field of track defect detection technology, and in particular to a method, apparatus, equipment and storage medium for detecting defects in regional ballastless tracks. Background Technology

[0002] Ballastless track is a critical infrastructure for modern high-speed and heavy-haul railways, and its structural health directly affects train safety. Currently, mainstream non-destructive testing methods rely on inspection vehicles integrating multiple sensors that automatically travel along the track and collect data. These sensors typically include: ground-penetrating radar (for detecting voids, moisture content, and density in the lower layer of the track slab), ultrasonic sensors (for detecting gaps between the track slab surface and interslabs), high-definition linear array cameras (for acquiring images of the track surface), and high-precision positioning and attitude determination systems (for providing accurate spatial coordinates and attitude for each set of detection data).

[0003] The conventional technical process involves the detection vehicle uploading massive amounts of raw data (especially large volumes of raw radar waveform data and high-definition image data) to a cloud data center in real time or near real time via mobile communication networks (such as 4G / 5G). In the cloud, powerful computing resources are used to run complex defect identification algorithms (such as deep learning models) to discover potential defects in the track structure.

[0004] However, the above-mentioned model faces severe challenges when the inspection mission covers areas with poor or completely interrupted network signal coverage, such as tunnels, mountains, and canyons. When the inspection vehicle passes through these areas, due to data upload channel congestion or a sharp drop in speed, the collected data quickly accumulates in the vehicle's limited storage buffer. Once the buffer is full, the system faces a dilemma: either stop inspection to wait for network recovery, resulting in a significant decrease in inspection efficiency and potential missed maintenance windows; or be forced to overwrite old data, causing permanent data loss. Both scenarios severely compromise the integrity of the inspection mission and the reliability of the data, making accurate assessment of track health difficult. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and storage medium for detecting defects in ballastless tracks in a region, aiming to solve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides a method for detecting defects in regional ballastless track, comprising the following steps:

[0007] Query the defect detection tasks for ballastless tracks within the target area; wherein, the defect detection task includes the track route information and defect detection period of the ballastless tracks to be detected;

[0008] Based on the route structure characteristics of the track route information and the historical defect database of ballastless track, the ballastless track to be inspected is divided into several inspection sub-segments. Based on the inspection item type and requirements of each inspection sub-segment, the inspection data upload requirements of each inspection sub-segment are generated.

[0009] Obtain a communication quality map of several track detection communication base stations within the target area, and determine the data upload capability of each detection sub-segment based on its location on the communication quality map.

[0010] Based on the upper limit of storage device capacity and candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track inspection vehicle allocated to the ballastless track to be inspected, the detection speed of the ballastless track inspection vehicle in each inspection sub-section is used as the decision variable. Considering the set of constraints constructed by the detection speed level, the amount of detection data storage and the defect detection time, as well as the optimization objective determined by the number of detection speed level switching, the regional ballastless track defect detection strategy is solved.

[0011] Based on the detection speed of each detection sub-segment in the regional ballastless track defect detection strategy, the ballastless track detection vehicle is controlled to perform ballastless track detection and upload detection data.

[0012] Optionally, query the defect detection task steps for the ballastless track to be inspected within the target area, specifically including:

[0013] After obtaining the track identifier of the ballastless track to be inspected, the track identifier is used to initiate a defect detection task query request to the ballastless track defect detection task management system.

[0014] The system receives the inspection window period and several track trajectory points for the ballastless track to be inspected from the ballastless track defect inspection task management system. The inspection window period is used as the defect inspection period, and the several track trajectory points are used as track route information to generate a defect inspection task.

[0015] Optionally, based on the route structure characteristics of the track route information and the historical defect database of ballastless track, the ballastless track to be inspected is divided into several inspection sub-segments. Based on the inspection item types and requirements of each inspection sub-segment, the steps for generating the inspection data upload requirements for each inspection sub-segment are generated, specifically including:

[0016] Extract several track trajectory points at fixed intervals from the track route information, use the position coordinates of two adjacent track trajectory points as the index of the corresponding detection sub-segment, and match several route structure features of each detection sub-segment in the route structure feature database;

[0017] Based on several route structure features, historical transport load features, and historical transport environment features of each detection sub-section, detection samples are constructed. The detection samples are then input into a pre-trained probability prediction model for the disease category of each detection sub-section to obtain the probability of each detection sub-section for each disease category.

[0018] The probability prediction model for the defect category of the detection sub-section is configured as follows: training samples are constructed by extracting route structure features, historical transport load features and historical transport environment features of several detection sub-sections from the historical defect database of ballastless track, and then input into a pre-built XGBoost classification model for training to obtain the probability prediction model for the defect category.

[0019] The target disease categories with probability values ​​exceeding preset values ​​are extracted. Based on the detection items and requirements corresponding to the target diseases, the data generation rate of the detection equipment corresponding to each detection item is summed to estimate the amount of data generated per unit detection length for each detection sub-segment.

[0020] Optionally, the route structure characteristics specifically include at least one or more combinations of ballastless track type, track bed structure, curve radius, gradient, and fastener type; the historical transport load characteristics specifically include at least one or more combinations of the number of trains passing through the track, train type, operating speed, train axle load, and cumulative transport volume; the historical transport environment characteristics specifically include at least one or more combinations of the average annual temperature, duration of extreme temperatures, average annual rainfall, surrounding soil type, and geological parameters.

[0021] Optionally, a communication quality map of several track detection communication base stations within the target area is obtained. Based on the location of each detection sub-segment in the communication quality map, the data upload capability of each detection sub-segment is determined. This step specifically includes:

[0022] The API interface provided by the rail communication operator is called to access the communication quality database of the rail detection communication base station and query the communication quality map of the target area; wherein, the communication quality map is configured to contain several area grids with expected data uplink transmission rate values;

[0023] The spatial trajectory of each detection sub-segment is overlaid and analyzed in the communication quality map. The average value of the expected uplink transmission rate of the data in the corresponding area grid of several spatial trajectory sampling points in the communication quality map is taken as the detection data upload capability of each detection sub-segment.

[0024] Optionally, based on the upper limit of storage device capacity and the candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track inspection vehicle allocated to the ballastless track to be inspected, and taking the inspection speed of the ballastless track inspection vehicle in each inspection sub-section as the decision variable, considering the constraint set constructed by the detection speed level, the amount of inspection data storage, and the defect detection time, as well as the optimization objective determined by the number of detection speed level switching, the steps of the regional ballastless track defect detection strategy are solved, specifically including:

[0025] Obtain the hardware attributes of the ballastless track inspection vehicle assigned to the ballastless track to be inspected; wherein, the hardware attributes include the upper limit of the storage capacity of the inspection data storage device and a candidate set of vehicle inspection speed levels containing several inspection speed levels;

[0026] The decision variable is the speed of the ballastless track inspection vehicle in each inspection sub-section. The first constraint is that the speed of the ballastless track inspection vehicle in each inspection sub-section falls into the candidate set of vehicle inspection speed levels. The second constraint is that the sum of the speed of the ballastless track inspection vehicle in each inspection sub-section and the detection time of the inspection sub-section determined by the length of the section line is less than the detection time of the defect detection period. The third constraint is that the value of the detection data storage at the beginning of each inspection sub-section, plus the detection time from the beginning to the current detection time, the detection length determined by the speed of the inspection sub-section, and the amount of data generated per unit detection length, minus the detection time from the beginning to the detection time and the amount of data uploaded determined by the detection data upload capability, is less than the upper limit of the storage device capacity.

[0027] With the goal of minimizing the sum of the number of times two adjacent sub-segments have different speed levels in several detection sub-segments, an optimization algorithm is used to solve the decision variables, and a regional ballastless track defect detection strategy is obtained, which includes the detection speed adopted by the ballastless track inspection vehicle in each detection sub-segment.

[0028] Optionally, based on the detection travel speed for each detection sub-segment in the regional ballastless track defect detection strategy, the ballastless track inspection vehicle is controlled to perform ballastless track inspection and data upload steps, specifically including:

[0029] The regional ballastless track defect detection strategy is loaded into the control terminal of the ballastless track inspection vehicle, and the GNSS positioning equipment configured on the ballastless track inspection vehicle is used to identify the inspection sub-section.

[0030] The ballastless track inspection vehicle is controlled to collect and upload ballastless track inspection data in each inspection sub-section according to the corresponding inspection travel speed in the regional ballastless track defect inspection strategy, so that the cloud can perform regional ballastless track defect inspection and analysis after receiving the ballastless track inspection data.

[0031] Furthermore, to achieve the above objectives, the present invention also provides a regional ballastless track defect detection device, comprising:

[0032] The query module is used to query defect detection tasks for ballastless tracks within a target area; wherein, the defect detection task includes track route information and defect detection time period of the ballastless track to be detected;

[0033] The segmentation module is used to divide the ballastless track to be inspected into several inspection sub-segments based on the route structure characteristics of the track route information and the historical defect database of ballastless track. Based on the inspection item type and requirements of each inspection sub-segment, the module generates the inspection data upload requirements for each inspection sub-segment.

[0034] The determination module is used to obtain a communication quality map of several track detection communication base stations in the target area, and determine the detection data upload capability of each detection sub-segment based on its location in the communication quality map.

[0035] The solution module is used to solve the regional ballastless track defect detection strategy based on the upper limit of storage device capacity and the candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track detection vehicles allocated to the ballastless track to be detected, taking the detection speed of the ballastless track detection vehicle in each detection sub-segment as the decision variable, considering the constraint set constructed by the detection speed level, the amount of detection data storage and the defect detection time, and the optimization objective determined by the number of detection speed level switching.

[0036] The control module is used to control the ballastless track inspection vehicle to perform ballastless track inspection and upload inspection data based on the inspection travel speed for each inspection sub-segment in the regional ballastless track defect inspection strategy.

[0037] In addition, to achieve the above objectives, the present invention also provides a regional ballastless track defect detection device, which includes: a memory, a processor, and a regional ballastless track defect detection program stored in the memory and executable on the processor. When the regional ballastless track defect detection program is executed by the processor, it implements the steps of the regional ballastless track defect detection method as described above.

[0038] In addition, to achieve the above objectives, the present invention also provides a storage medium storing a regional ballastless track defect detection program, which, when executed by a processor, implements the steps of the above-described regional ballastless track defect detection method.

[0039] The beneficial effects of this invention are as follows: It proposes a method, device, equipment, and storage medium for detecting defects in ballastless track in a region. By querying the defect detection tasks of the ballastless track to be detected, and based on the historical defect database of ballastless track, the ballastless track to be detected is divided into several detection sub-segments. The detection data upload requirements for each detection sub-segment are generated. Based on the communication quality map, the detection data upload capability of each detection sub-segment is determined. Then, taking the speed of the detection vehicle as the core, and combining hardware constraints and gear switching optimization objectives, the optimal detection strategy is solved. This guides the ballastless track detection vehicle to perform ballastless track detection driving, detection data upload, and cloud-based defect detection analysis. Therefore, by optimizing the inspection speed of ballastless track inspection vehicles in different inspection sub-sections, this invention ensures that inspections are completed within the maintenance window, while completely solving the problem of data accumulation in areas with poor network conditions such as tunnels and mountainous areas, which leads to the cessation of inspections or the overwriting of old data. This ensures the integrity of inspection tasks and the reliability of data, reduces wear and tear and accuracy loss on inspection equipment caused by gear switching, helps the cloud to quickly analyze and output defect detection results, provides timely basis for track maintenance, and reduces traffic safety risks. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention;

[0041] Figure 2 This is a schematic flowchart of an embodiment of the method for detecting defects in ballastless track in the region according to the present invention;

[0042] Figure 3 This is a structural block diagram of a regional ballastless track defect detection device according to an embodiment of the present invention. Detailed Implementation

[0043] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0045] like Figure 1 As shown, Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.

[0046] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU; a communication bus 1002; a user interface 1003; a network interface 1004; and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0047] Those skilled in the art will understand that Figure 1 The structure of the device shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0048] like Figure 1 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a regional ballastless track defect detection program.

[0049] exist Figure 1 In the terminal shown, network interface 1004 is mainly used to connect to the backend server and communicate data with it; user interface 1003 is mainly used to connect to the client (user terminal) and communicate data with it; while processor 1001 can be used to call the regional ballastless track defect detection program stored in memory 1005 and perform the following operations:

[0050] Query the defect detection tasks for ballastless tracks within the target area; wherein, the defect detection task includes the track route information and defect detection period of the ballastless tracks to be detected;

[0051] Based on the route structure characteristics of the track route information and the historical defect database of ballastless track, the ballastless track to be inspected is divided into several inspection sub-segments. Based on the inspection item type and requirements of each inspection sub-segment, the inspection data upload requirements of each inspection sub-segment are generated.

[0052] Obtain a communication quality map of several track detection communication base stations within the target area, and determine the data upload capability of each detection sub-segment based on its location on the communication quality map.

[0053] Based on the upper limit of storage device capacity and candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track inspection vehicle allocated to the ballastless track to be inspected, the detection speed of the ballastless track inspection vehicle in each inspection sub-section is used as the decision variable. Considering the set of constraints constructed by the detection speed level, the amount of detection data storage and the defect detection time, as well as the optimization objective determined by the number of detection speed level switching, the regional ballastless track defect detection strategy is solved.

[0054] Based on the detection speed of each detection sub-segment in the regional ballastless track defect detection strategy, the ballastless track detection vehicle is controlled to perform ballastless track detection and upload detection data.

[0055] The specific embodiments of the present invention applied to the device are basically the same as the embodiments of the ballastless track defect detection method in the application area described below, and will not be repeated here.

[0056] This invention provides a method for detecting defects in regional ballastless track, referring to... Figure 2 , Figure 2 This is a schematic flowchart illustrating an embodiment of the regional ballastless track defect detection method of the present invention.

[0057] In this embodiment, a method for detecting defects in regional ballastless track includes the following steps:

[0058] S100: Query the defect detection tasks for the ballastless track to be detected within the target area; wherein, the defect detection task includes the track route information and defect detection period of the ballastless track to be detected;

[0059] S200: Based on the route structure characteristics of the track route information and the historical defect database of ballastless track, the ballastless track to be inspected is divided into several inspection sub-segments. Based on the inspection item type and requirements of each inspection sub-segment, the inspection data upload requirements of each inspection sub-segment are generated.

[0060] S300: Obtain a communication quality map of several track detection communication base stations within the target area, and determine the data upload capability of each detection sub-segment based on its location on the communication quality map;

[0061] S400: Based on the upper limit of storage device capacity and candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track inspection vehicle allocated to the ballastless track to be inspected, the detection speed of the ballastless track inspection vehicle in each inspection sub-segment is used as the decision variable. Considering the set of constraints constructed by the detection speed level, the amount of detection data storage and the defect detection time, as well as the optimization objective determined by the number of detection speed level switching, the regional ballastless track defect detection strategy is solved.

[0062] S500: Based on the detection travel speed for each detection sub-segment in the regional ballastless track defect detection strategy, control the ballastless track detection vehicle to perform ballastless track detection and upload detection data.

[0063] It should be noted that traditional methods face severe challenges when inspection missions cover areas with poor or completely interrupted network signal coverage, such as tunnel complexes, mountainous areas, and canyons. When inspection vehicles pass through these areas, due to data upload channel congestion or a sharp drop in speed, the collected data quickly accumulates in the vehicle's limited storage buffer. Once the buffer is full, the system faces a dilemma: either stop inspection to wait for network recovery, resulting in a significant decrease in inspection efficiency and potential missed maintenance windows; or be forced to overwrite old data, causing permanent data loss. Both scenarios severely compromise the integrity of the inspection mission and the reliability of the data, making accurate assessment of track health difficult.

[0064] To address the aforementioned issues, this embodiment queries the defect detection tasks for the ballastless track to be inspected, divides the track route into inspection sub-segments based on structural characteristics, generates data upload requirements, determines the upload capacity of each segment based on a communication quality map, optimizes the inspection speed based on the hardware attributes and constraints of the inspection vehicle, and finally controls the vehicle to execute inspections and data uploads, generating a regional ballastless track defect detection strategy. By optimizing the inspection speed of the ballastless track inspection vehicle in different inspection sub-segments, while ensuring that inspections are completed within the maintenance window, it completely solves the problem of data accumulation in areas with poor network connectivity, such as tunnels and mountainous areas, leading to inspection stoppages or the overwriting of old data. This ensures the integrity of the inspection tasks and the reliability of the data, reduces wear and tear and accuracy degradation on inspection equipment caused by gear switching, helps the cloud quickly analyze and output defect detection results, provides timely information for track maintenance, and reduces traffic safety risks.

[0065] In a preferred embodiment, the steps for detecting defects in the ballastless track within the target area specifically include:

[0066] S110: After obtaining the track identifier of the ballastless track to be inspected, use the track identifier to initiate a defect detection task query request to the ballastless track defect detection task management system.

[0067] S120: Receive the inspection window period and several track trajectory points for the ballastless track to be inspected from the ballastless track defect inspection task management system, take the inspection window period as the defect inspection period, take the several track trajectory points as track route information, and generate a defect inspection task.

[0068] In this embodiment, firstly, a unique identifier (such as track line number, mileage segment code, etc.) of the ballastless track to be inspected is obtained. Using this identifier as an index, a defect inspection task query request is sent to the ballastless track defect inspection task management system. Then, the response data returned by the task management system is received, from which two core pieces of information are extracted: the inspection window period (i.e., the time period during which inspection work is allowed, which must avoid train running hours) and several track trajectory points (such as coordinate points every 50m, used to determine the specific direction of the track). Finally, the extracted inspection window period is defined as the defect inspection period, and the set of track trajectory points is defined as the track route information. The two are integrated to form a complete defect inspection task, providing basic data for subsequent section division and inspection execution.

[0069] As is easily understood, this embodiment initiates a query request to the ballastless track defect detection task management system for the track identifier of the ballastless track to be inspected, receives the inspection window period and track trajectory points issued by the system, and generates a complete defect detection task, which can ensure the accuracy and relevance of the inspection task information: (1) By querying through a unique track identifier, the task is avoided from being confused with other tracks; (2) Clearly defining the window period can prevent the inspection operation from conflicting with train operation and ensure construction safety; (3) Track trajectory points provide accurate spatial basis for subsequent section division and positioning, avoiding detection deviations caused by ambiguous route information.

[0070] In a preferred embodiment, based on the route structure characteristics of the track route information and the historical defect database of ballastless track, the ballastless track to be inspected is divided into several inspection sub-segments. Based on the inspection item types and requirements of each inspection sub-segment, a data upload requirement step for each inspection sub-segment is generated, specifically including:

[0071] S210: Extract several track trajectory points at fixed intervals from the track route information, use the position coordinates of two adjacent track trajectory points as the index of the corresponding detection sub-segment, and match several route structure features of each detection sub-segment in the route structure feature database.

[0072] S220: Construct detection samples based on several route structure features, historical transport load features, and historical transport environment features of each detection sub-section. Input the detection samples into the pre-trained detection sub-section disease category probability prediction model to obtain the probability of each detection sub-section for each disease category.

[0073] The probability prediction model for the defect category of the detection sub-section is configured as follows: training samples are constructed by extracting route structure features, historical transport load features and historical transport environment features of several detection sub-sections from the historical defect database of ballastless track, and then input into a pre-built XGBoost classification model for training to obtain the probability prediction model for the defect category.

[0074] S230: Extract the target disease categories whose probability values ​​exceed the preset value, and estimate the amount of data generated per unit detection length for each detection sub-segment by summing the data generation rate of the detection equipment corresponding to each detection item, based on the detection items and requirements corresponding to the target disease.

[0075] In this embodiment, firstly, trajectory points at fixed intervals are extracted from the track route information. The position coordinates of two adjacent trajectory points are used as the unique index of the corresponding detection sub-segment (e.g., the index of the "K30.000-K30.050" segment is the coordinates of two points). Then, using the segment index as the keyword, the structural features of the segment are queried and matched in the route structure feature database. Next, the route structure features, historical transport load features, and historical transport environment features of each segment are integrated into a detection sample, which is then input into a pre-trained XGBoost classification model (the model training data comes from similar features in the ballastless track historical defect database). The system generates a sample of the target defects and outputs the probability of each defect type for that section (e.g., 30% probability of CA mortar voids and 15% probability of track slab cracks). Finally, it extracts the target defect types with probabilities exceeding a preset value (e.g., 20%). Based on the detection items corresponding to the target defects (e.g., radar detection for CA mortar voids) and the requirements (data sampling frequency, which is usually proportional to the detection travel speed, i.e., the number of data samples per unit detection length changes with the detection travel speed, while the total amount of data remains unchanged), it sums up the data generation rates of each detection item and calculates the amount of data generated per unit detection length of that section (i.e., the data upload requirement).

[0076] Therefore, this embodiment extracts track trajectory points at fixed intervals based on track route information, matches route structure features using adjacent trajectory points as indices, constructs samples by combining historical load and environmental features, obtains the probability of defects through an XGBoost prediction model, and finally estimates the amount of detection data per unit length for each detection sub-segment, generating data upload requirements. This achieves accurate division of detection sub-segments and on-demand uploading. Through historical features and the XGBoost model, high-risk defect segments are accurately identified, avoiding resource waste caused by indiscriminate detection. Estimating data volume by detection item provides a quantitative basis for subsequent matching of communication upload capabilities and optimization of detection speed, preventing data accumulation or loss due to mismatch between data volume and upload capacity.

[0077] For example, the route structure characteristics specifically include at least one or more combinations of ballastless track type, track bed structure, curve radius, gradient, and fastener type; the historical transport load characteristics specifically include at least one or more combinations of the number of trains passing through the track, train type, operating speed, train axle load, and cumulative transport volume; the historical transport environment characteristics specifically include at least one or more combinations of the average annual temperature, duration of extreme temperatures, average annual rainfall, surrounding soil type, and geological parameters.

[0078] In practical applications, by clearly defining the specific dimensions of route structure characteristics, historical transport load characteristics, and historical transport environment characteristics, standardized feature inputs are provided for the construction of detection samples and the prediction of defect probabilities. Specifically, route structure characteristics are used to focus on the inherent properties of the track itself, historical transport load characteristics are used to focus on the dynamic effects of train operation on the track, and historical transport environment characteristics are used to focus on the long-term impact of the natural and surrounding environment on the track. Thus, comprehensive and standardized feature inputs are provided for the defect probability prediction model, improving the accuracy of the XGBoost model in predicting defect probabilities.

[0079] In a preferred embodiment, the step of acquiring a communication quality map of several track detection communication base stations within the target area, and determining the detection data upload capability of each detection sub-segment based on its location on the communication quality map, specifically includes:

[0080] S310: Call the API interface provided by the rail communication operator to access the communication quality database of the rail detection communication base station and query the communication quality map of the target area; wherein, the communication quality map is configured to contain several area grids with expected data uplink transmission rate values;

[0081] S320: The spatial trajectory of each detection sub-segment is overlaid and analyzed in the communication quality map, and the average value of the expected data uplink transmission rate of the corresponding area grid of several spatial trajectory sampling points in the communication quality map is taken as the detection data upload capability of each detection sub-segment.

[0082] In this embodiment, the standardized API interface provided by the railway private network's track communication operator is called to access the communication quality database of its track detection communication base station and query the communication quality map covering the target area. This map uses regional grids as the basic unit (e.g., 10m×10m), and each grid is marked with the expected value of the data uplink transmission rate at the corresponding location (e.g., 10Mbps, 50Mbps). The spatial trajectory of each detection sub-segment (composed of multiple sampling points within the segment, e.g., one sampling point every 10m) is spatially overlaid with the communication quality map for analysis. The expected value of the uplink rate of the grid where each trajectory sampling point is located is extracted, and the average uplink rate of all trajectory sampling points in the segment is calculated. This average value is used as the detection data upload capability of the detection sub-segment (i.e., the average uplink rate that the segment can stably achieve).

[0083] Therefore, this embodiment accesses the communication quality database by calling the API interface of the rail communication operator to obtain a communication quality map containing uplink rate grids. By overlaying the spatial trajectory of the detection sub-segment, the average uplink rate of the trajectory sampling points is calculated to determine the data upload capability of each segment. This accurately quantifies the communication upload capability of each detection sub-segment, providing a crucial basis for subsequent speed optimization. It avoids the mismatch between detection speed and upload capability caused by unknown upload rates (e.g., high-speed acquisition generates a large amount of data but the upload rate is low, leading to data accumulation). By overlaying the gridded map with the trajectory, the spatial accuracy of the upload capability calculation is ensured, making it particularly suitable for areas with large communication quality fluctuations, such as tunnels and alternating open-air areas.

[0084] In a preferred embodiment, based on the upper limit of storage device capacity and the candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track inspection vehicle allocated to the ballastless track to be inspected, and taking the detection speed of the ballastless track inspection vehicle in each inspection sub-segment as the decision variable, considering the constraint set constructed by the detection speed level, the amount of detection data storage, and the defect detection time, as well as the optimization objective determined by the number of detection speed level switching, the steps for solving the regional ballastless track defect detection strategy are specifically included:

[0085] S410: Obtain the hardware attributes of the ballastless track detection vehicle assigned to the ballastless track to be detected; wherein, the hardware attributes include the upper limit of the storage capacity of the detection data storage device and a candidate set of vehicle detection speed levels containing several detection speed levels.

[0086] S420: The decision variable is the detection speed of the ballastless track inspection vehicle in each inspection sub-section; the first constraint is that the detection speed of the ballastless track inspection vehicle in each inspection sub-section falls into the candidate set of vehicle detection speed gears; the second constraint is that the sum of the detection speed of the ballastless track inspection vehicle in each inspection sub-section and the detection time of the inspection sub-section determined by the section length of the line in each sub-section is less than the detection time of the defect detection period; the third constraint is that the detection data storage amount of the ballastless track inspection vehicle at the beginning of each inspection sub-section, plus the detection time from the beginning to the current detection time and the detection length determined by the detection speed of the inspection sub-section and the amount of data generated per unit detection length, minus the detection time from the beginning to the detection time and the amount of uploaded data determined by the detection data upload capability, is less than the upper limit of the storage device capacity.

[0087] S430: Taking the minimum sum of the number of times the speed levels of two adjacent sub-segments in several detection sub-segments are different as the optimization objective, the optimization algorithm is used to solve the decision variables to obtain the regional ballastless track defect detection strategy, which includes the detection travel speed adopted by the ballastless track detection vehicle in each detection sub-segment.

[0088] In this embodiment, the hardware parameters of the detection vehicle allocated to the track to be detected are first obtained, including the upper limit of the storage device capacity (i.e., the maximum amount of data that the on-board cache can store) and the candidate set of vehicle detection speed gears (e.g., {5km / h, 10km / h, 15km / h}, the vehicle can only drive at the preset gear). Then, the decision variables and constraints are constructed: (1) Decision variable: the detection speed of the detection vehicle in each detection sub-section; (2) First constraint: the decision variable must belong to the candidate set of vehicle detection speed gears (e.g., the speed can only be 5 / 10 / 15km / h, not 7km / h); (3) Second constraint: the sum of the detection time (segment length / detection speed) of all sub-sections must be less than the disease detection period (e.g., the window period of 4 hours); (4) Third constraint: within each sub-section, the current storage amount + data generation amount - data upload amount < storage device capacity limit; where, data generation amount = segment length × data amount per unit length, data upload amount = detection time × upload capacity. Finally, with the goal of minimizing the sum of the number of times the speed levels of all adjacent sub-segments are different (e.g., 5km / h for segment 1 and 5km / h for segment 2 with no switching; 5km / h for segment 2 and 10km / h for segment 3 with 1 switching), an optimization algorithm (such as a genetic algorithm) is used to solve the decision variables to obtain the optimal detection strategy that includes the detection speed of each sub-segment.

[0089] Therefore, this embodiment uses the vehicle's speed in each sub-segment as the decision variable, and combines the vehicle's hardware attributes (storage capacity, speed gear) to construct constraints (gear compliance, time limit, storage not exceeding limits), with the goal of reducing the number of speed gear switching. The optimal detection strategy is obtained by solving through an optimization algorithm. Under the premise of ensuring complete detection, no data loss, and time compliance, the detection process is efficient and stable: (1) The constraints ensure that the detection task is completed within the window period and that the storage does not exceed the limit, avoiding data loss or task interruption; (2) The optimization goal is to reduce gear switching, reduce vehicle mechanical wear (such as wear on the transmission system from frequent gear changes), and at the same time, reduce the frequency of vehicle speed changes as much as possible, which can reduce the frequent switching of different detection devices, reduce the switching accuracy error of detection devices and the load on the controller.

[0090] In a preferred embodiment, based on the detection travel speed for each detection sub-segment in the regional ballastless track defect detection strategy, the ballastless track detection vehicle is controlled to perform ballastless track detection and detection data upload steps, specifically including:

[0091] S510: Load the regional ballastless track defect detection strategy into the control terminal of the ballastless track inspection vehicle, and identify the inspection sub-segment through the GNSS positioning equipment configured on the ballastless track inspection vehicle;

[0092] S520: Controls the ballastless track inspection vehicle to collect and upload ballastless track inspection data in each inspection sub-section according to the corresponding inspection travel speed in the regional ballastless track defect inspection strategy, so that the cloud can perform regional ballastless track defect inspection and analysis after receiving the ballastless track inspection data.

[0093] In this embodiment, the optimized detection strategy is loaded into the detection vehicle control terminal. The GNSS positioning device identifies the current detection sub-segment, and the vehicle is controlled to perform detection data collection and synchronous uploading at the speed specified by the strategy. This ensures that the cloud receives the data and completes the defect analysis. On one hand, GNSS positioning ensures that the vehicle accurately matches the segment speed, avoiding speed errors caused by positioning deviations. On the other hand, synchronous collection and uploading reduces the pressure on onboard storage, especially in segments with poor network connectivity. Collecting data at the optimized speed balances data generation and uploading, preventing data backlog, and thus enabling real-time cloud analysis to shorten the defect identification cycle and facilitate rapid development of repair plans.

[0094] Reference Figure 3 , Figure 3 This is a structural block diagram of an embodiment of the regional ballastless track defect detection device of the present invention.

[0095] like Figure 3 As shown, the regional ballastless track defect detection device proposed in this embodiment of the invention includes:

[0096] The query module 10 is used to query the defect detection tasks of the ballastless track to be detected within the target area; wherein, the defect detection task includes the track route information and defect detection period of the ballastless track to be detected;

[0097] The segmentation module 20 is used to divide the ballastless track to be inspected into several inspection sub-segments based on the route structure characteristics of the track route information and the historical defect database of ballastless track, and to generate the inspection data upload requirements for each inspection sub-segment based on the inspection item type and requirements of each inspection sub-segment.

[0098] The determination module 30 is used to obtain a communication quality map of several track detection communication base stations in the target area, and determine the detection data upload capability of each detection sub-segment based on the location of each detection sub-segment in the communication quality map.

[0099] The solution module 40 is used to solve the regional ballastless track defect detection strategy based on the upper limit of storage device capacity and the candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track detection vehicle allocated to the ballastless track to be detected, taking the detection speed of the ballastless track detection vehicle in each detection sub-segment as the decision variable, considering the constraint set constructed by the detection speed level, the amount of detection data storage and the defect detection time, and the optimization objective determined by the number of detection speed level switching.

[0100] The control module 50 is used to control the ballastless track inspection vehicle to perform ballastless track inspection and upload inspection data based on the inspection travel speed for each inspection sub-segment in the regional ballastless track defect inspection strategy.

[0101] Other embodiments or specific implementations of the regional ballastless track defect detection device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0102] Furthermore, the present invention also proposes a regional ballastless track defect detection device, which includes: a memory, a processor, and a regional ballastless track defect detection program stored in the memory and executable on the processor. When the regional ballastless track defect detection program is executed by the processor, it implements the steps of the regional ballastless track defect detection method as described above.

[0103] The specific implementation method of the regional ballastless track defect detection equipment in this application is basically the same as the embodiments of the above-mentioned regional ballastless track defect detection methods, and will not be repeated here.

[0104] Furthermore, this invention also proposes a readable storage medium, which includes a computer-readable storage medium storing a regional ballastless track defect detection program. The readable storage medium may be... Figure 1 The memory 1005 in the terminal can also be at least one of ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk, etc. The readable storage medium includes several instructions to cause a regional ballastless track defect detection device with a processor to execute the regional ballastless track defect detection method described in various embodiments of the present invention.

[0105] The specific implementation methods in the readable storage medium of this application are basically the same as the embodiments of the above-described method for detecting defects in ballastless track in the region, and will not be described again here.

[0106] It is understood that in the description of this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first embodiment to Nth embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0107] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0108] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0109] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0110] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for detecting defects in regional ballastless track, characterized in that, Includes the following steps: Query the defect detection tasks for ballastless tracks within the target area; wherein, the defect detection task includes the track route information and defect detection period of the ballastless tracks to be detected; Based on the route structure characteristics of the track route information and the historical defect database of ballastless track, the ballastless track to be inspected is divided into several inspection sub-segments. Based on the inspection item type and requirements of each inspection sub-segment, the inspection data upload requirements for each inspection sub-segment are generated. The inspection data upload requirements are configured as the amount of data generated per unit inspection length. A communication quality map of several track detection communication base stations within a target area is obtained. Based on the location of each detection sub-segment in the communication quality map, the detection data upload capability of each detection sub-segment is determined. The detection data upload capability is configured as the average value of the expected uplink transmission rate of the data in the corresponding area grid of several spatial trajectory sampling points in the communication quality map. Based on the upper limit of storage capacity and the candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track inspection vehicles allocated to the ballastless track to be inspected, and taking the inspection speed of the ballastless track inspection vehicles in each inspection sub-segment as the decision variable, the optimization objective determined by the set of constraints constructed from the detection speed level, the amount of inspection data stored, and the defect detection time, as well as the number of detection speed level switching times, the regional ballastless track defect detection strategy is solved; specifically including: Obtain the hardware attributes of the ballastless track inspection vehicle assigned to the ballastless track to be inspected; wherein, the hardware attributes include the upper limit of the storage capacity of the inspection data storage device and a candidate set of vehicle inspection speed levels containing several inspection speed levels; The decision variable is the speed of the ballastless track inspection vehicle in each inspection sub-section. The first constraint is that the speed of the ballastless track inspection vehicle in each inspection sub-section falls into the candidate set of vehicle inspection speed levels. The second constraint is that the sum of the speed of the ballastless track inspection vehicle in each inspection sub-section and the detection time of the inspection sub-section determined by the length of the section line is less than the detection time of the defect detection period. The third constraint is that the value of the detection data storage at the beginning of each inspection sub-section, plus the detection time from the beginning to the current detection time, the detection length determined by the speed of the inspection sub-section, and the amount of data generated per unit detection length, minus the detection time from the beginning to the detection time and the amount of data uploaded determined by the detection data upload capability, is less than the upper limit of the storage device capacity. With the goal of minimizing the sum of the number of times the speed levels of two adjacent sub-segments in several detection sub-segments are different, an optimization algorithm is used to solve the decision variables to obtain a regional ballastless track defect detection strategy consisting of the detection travel speed adopted by the ballastless track detection vehicle in each detection sub-segment. Based on the detection speed of each detection sub-segment in the regional ballastless track defect detection strategy, the ballastless track detection vehicle is controlled to perform ballastless track detection and upload detection data.

2. The method for detecting defects in regional ballastless track as described in claim 1, characterized in that, The steps for detecting defects in ballastless track within the target area include: After obtaining the track identifier of the ballastless track to be inspected, the track identifier is used to initiate a defect detection task query request to the ballastless track defect detection task management system. The system receives the inspection window period and several track trajectory points for the ballastless track to be inspected from the ballastless track defect inspection task management system. The inspection window period is used as the defect inspection period, and the several track trajectory points are used as track route information to generate a defect inspection task.

3. The method for detecting defects in regional ballastless track as described in claim 1, characterized in that, Based on the route structure characteristics of the track route information and the historical defect database of ballastless track, the ballastless track to be inspected is divided into several inspection sub-segments. Based on the inspection item types and requirements of each inspection sub-segment, the data upload requirements for each inspection sub-segment are generated, specifically including: Extract several track trajectory points at fixed intervals from the track route information, use the position coordinates of two adjacent track trajectory points as the index of the corresponding detection sub-segment, and match several route structure features of each detection sub-segment in the route structure feature database; Based on several route structure features, historical transport load features, and historical transport environment features of each detection sub-section, detection samples are constructed. The detection samples are then input into a pre-trained probability prediction model for the disease category of each detection sub-section to obtain the probability of each detection sub-section for each disease category. The probability prediction model for the defect category of the detection sub-section is configured as follows: training samples are constructed by extracting route structure features, historical transport load features and historical transport environment features of several detection sub-sections from the historical defect database of ballastless track, and then input into a pre-built XGBoost classification model for training to obtain the probability prediction model for the defect category. The target disease categories with probability values ​​exceeding preset values ​​are extracted. Based on the detection items and requirements corresponding to the target diseases, the data generation rate of the detection equipment corresponding to each detection item is summed to estimate the amount of data generated per unit detection length for each detection sub-segment.

4. The method for detecting defects in regional ballastless track as described in claim 3, characterized in that, The route structure characteristics specifically include at least one or more combinations of ballastless track type, track bed structure, curve radius, gradient, and fastener type; the historical transport load characteristics specifically include at least one or more combinations of the number of trains passing through the track, train type, operating speed, train axle load, and cumulative transport volume; the historical transport environment characteristics specifically include at least one or more combinations of the average annual temperature, duration of extreme temperatures, average annual rainfall, surrounding soil type, and geological parameters.

5. The method for detecting defects in regional ballastless track as described in claim 1, characterized in that, The steps for obtaining a communication quality map of several track detection communication base stations within a target area, and determining the data upload capability of each detection sub-segment based on its location on the communication quality map, specifically include: The API interface provided by the rail communication operator is called to access the communication quality database of the rail detection communication base station and query the communication quality map of the target area; wherein, the communication quality map is configured to contain several area grids with expected data uplink transmission rate values; The spatial trajectory of each detection sub-segment is overlaid and analyzed in the communication quality map. The average value of the expected uplink transmission rate of the data in the corresponding area grid of several spatial trajectory sampling points in the communication quality map is taken as the detection data upload capability of each detection sub-segment.

6. The method for detecting defects in regional ballastless track as described in claim 1, characterized in that, Based on the detection travel speed for each detection sub-segment in the aforementioned regional ballastless track defect detection strategy, the ballastless track detection vehicle is controlled to perform ballastless track detection and detection data upload steps, specifically including: The regional ballastless track defect detection strategy is loaded into the control terminal of the ballastless track inspection vehicle, and the GNSS positioning equipment configured on the ballastless track inspection vehicle is used to identify the inspection sub-section. The ballastless track inspection vehicle is controlled to collect and upload ballastless track inspection data in each inspection sub-section according to the corresponding inspection travel speed in the regional ballastless track defect inspection strategy, so that the cloud can perform regional ballastless track defect inspection and analysis after receiving the ballastless track inspection data.

7. A regional ballastless track defect detection device, characterized in that, The method for detecting defects in ballastless track as described in any one of claims 1-6 includes: The query module is used to query defect detection tasks for ballastless tracks within a target area; wherein, the defect detection task includes track route information and defect detection time period of the ballastless track to be detected; The segmentation module is used to divide the ballastless track to be inspected into several inspection sub-segments based on the route structure characteristics of the track route information and the historical defect database of ballastless track. Based on the inspection item type and requirements of each inspection sub-segment, the module generates the inspection data upload requirements for each inspection sub-segment. The determination module is used to obtain a communication quality map of several track detection communication base stations in the target area, and determine the detection data upload capability of each detection sub-segment based on its location in the communication quality map. The solution module is used to solve the regional ballastless track defect detection strategy based on the upper limit of storage device capacity and the candidate set of vehicle detection speed levels in the hardware attributes of the ballastless track detection vehicles allocated to the ballastless track to be detected, taking the detection speed of the ballastless track detection vehicle in each detection sub-segment as the decision variable, considering the constraint set constructed by the detection speed level, the amount of detection data storage and the defect detection time, and the optimization objective determined by the number of detection speed level switching. The control module is used to control the ballastless track inspection vehicle to perform ballastless track inspection and upload inspection data based on the inspection travel speed for each inspection sub-segment in the regional ballastless track defect inspection strategy.

8. A regional ballastless track defect detection device, characterized in that, The regional ballastless track defect detection device includes: a memory, a processor, and a regional ballastless track defect detection program stored in the memory and executable on the processor. When the regional ballastless track defect detection program is executed by the processor, it implements the steps of the regional ballastless track defect detection method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium stores a regional ballastless track defect detection program, which, when executed by a processor, implements the steps of the regional ballastless track defect detection method as described in any one of claims 1 to 6.

Images